We had a freepbx 14 instance running previously, it got compromised, we wiped that instance and installed 17, so far so good but running into issues with the trunk config. the info from the trunk provider I have was “SIP Customer IP Block was 10.0.20.138/29” “SIP Core Network 172.45.31.200.and 219 for a backup”. I have eth0 set as the local lan the phones connect to and eth1 the 10.0.20.138 no gateway, I had added a route " route add -net 172.45.31.0 netmask 255.255.255.0 gw 10.0.20.138 dev eth1 "
What did I miss, the trunk wont connect. Sorry if this is a duplicate I quickly looked but couldn’t find the same scenario.
Update, so we got the trunk to connect with a very similar command, but new issue now the calls drop after 30 seconds even if leaving a voicemail, though I can stay in the IVR indefinitely. I thought it was the external IP part as I was not sure what should be in there, I have a 192, network, phones, and a 10. network that goes out directly to the trunk external via the ISP modem, it auto detected the wan IP but that as well as the 10. network ip did not work, same behavior.
Yes actually it went from sort of working to not again, so temporarily moved them to a cloud server. But i still need to get this working still, the part that bothers me is i even flattened the 17 version and put 14 back on with the same config that worked before…it connects to the trunk, but no calls pass through. I am getting from the isp we are handing them the wrong reply ip, but i have changed the external to the vlan ip and i ran a route command to pass traffic to the vlan subnet from the phone subnet the phones could call internal extensions also.
Which IP are you replying with that is wrong?
The isp states they want to see the 10. Network, i set the reply to the ip of the sever also, i tried the 10. Gateway ip also, as my notes before had the route stating the sip subnet to the gateway 10. Network. But i dont have a reaponse from them now after i changed it back from the wan ip it auto filled.
Could you confirm that you are using chan_pjsip, as that has much better support for the sort of half multi-homed situation that you seem to be describing, than had the old, chan_sip.
Hi, sorry for the delay kids events all weekend. I have tried both, though currently on the 14 version it is running on chan_sip. I can post the “requirements” from the but other than the odd (out of ordinary) RTP ports, the rest looks to be fairly straight forward.
Why don’t you post the network config for the 14 box. The routes, et al.
They should be controlling the remote ports, and not constraining your local ports, so I’m not sure how they could be specifying “out of ordinary” ports. There are no reserved ports for RTP, although there are reserved ones for other purposes, which need to be avoided.